Token cost per 10-job run
gpt-4o-miniSavings come from embedding the resume in the system prompt. OpenAI's automatic caching covers identical system-prompt prefixes — all 10 jobs share the same cached profile.
Tutorials
Follow setup in real execution order.
Each card covers one stage of project adoption, from first clone to daily automation.
Workflow
How one job row becomes generated documents.
Each step processes independently — one failure does not stop the rest.
Cost Optimization
~$0.002 per 10-job run.
Compact resume profiles + automatic OpenAI system-prompt caching keep costs low. The profile is cached after the first call and reused at 50 % cost for every subsequent job in the same run.
Cost controls
VariablesMAX_JOBS_PER_RUN
Hard cap per workflow execution. Default: 10.
OPENAI_MODEL
Default gpt-4o-mini — upgrade to gpt-4o only if output quality is insufficient.
JD character limit
Job descriptions capped at 4 000 chars (cover letter) / 3 000 chars (email) before API calls.
max_tokens
600 per cover letter · 400 per recruiter email — conservative limits prevent runaway cost.
Reference
Relevant data in one place.
Defaults, status values, folders, and test coverage summarized for quick lookup.
Core Configuration
Env VarsSpreadsheet Status Flow
State MachineProject Directories
File LayoutTest Coverage
Current SuiteCommand Deck